This application addresses the provocative question (PQ 17): Since current methods to assess potential cancer treatments are cumbersome, expensive, and often inaccurate, can we develop other methods to rapidly test interventions for cancer treatment or prevention? We will develop and apply a microfluidic system that monitors cell growth by measuring the mass of individual cells with unprecedented precision and speed before and after delivery of a particular drug. By analyzing the growth response from hundreds of cancer cells in only a few hours, we will determine if the differential growth response can ultimately be predictive of patient response and outcomes. Our application has broad implications for all types of cancer but our specific focus will be on glioblastoma (GBM) due to the advantageous model systems available. Currently, there are delays of months between initial drug administration and first assessment of response in patients. Ideally, predictions would happen in real- time, perhaps even while the patient remains in the operating room. Existing efforts in personalized medicine to assess the potential efficacy of a treatment are focused on identifying genomic and epigenetic predictors of response. However, such approaches have limitations since prediction of therapeutic response is based on indirect evidence from previously studied patient cohorts. Success is more likely to come from a combination of genetics and direct measures of patient tumor cell growth in response to drugs. Should such individualized tumor response testing be validated, the impact on cancer diagnostics and treatment would be profound. Here we aim to accomplish this through a combination of three innovative components: i) the suspended microchannel resonator (SMR) growth monitoring system which can monitor the mass of single live tumor cells with a precision near 0.01%, ii) a cohort of 20 patient-derived glioblastoma cell lines that recapitulate the tumor from which they were created based on comprehensive genomic, mutation, and expression profiling analysis, and iii) immediate access to fresh patient tissue samples through collaboration with the clinical DFCI/BWH Neurooncology Program. Our specific aims are: Aim One: Measure long-term growth properties from multiple established patient-derived cell lines and evaluate effects of combined drug therapy. Aim Two: Develop a high throughput platform for monitoring the mass of several hundred cells at a time Aim Three: Measure instantaneous growth rates (iGR) from multiple established patient-derived cell lines. Aim Four: Measure instantaneous growth rates (iGR) and long-term growth properties from patient samples. The results of this research are expected to lead to novel diagnostic tools fr drug selection and personalized predictions of patient response to targeted therapy.